South African Avocado Growers’ Association Yearbook 1993. 16:5-8
CUKE DEVELOPMENT IN CONSISTENTLY LOW PRODUCING TREES
OF THE 'FUERTE' AVOCADO WITH SPECIAL REFERENCE TO SEED
ABORTION
E.M.A. STEYN1 , P.J. ROBBERTSE1 AND D.G. SMITH2
Margaretha Mes Institute, Department of Botany, University of Pretoria, Pretoria 0002,
RSA1 Merensky Technological Services, P O Box 14, Duiwelskloof 08352
A full length paper on this topic was published in "Sexual Plant Reproduction" Vol 6
(1993, in press) by the same authors and is obtainable from the second author.
INTRODUCTION
The mechanism of fruit development is normally set in motion by two consecutive
stimuli. Auxins and gibberellins produced during pollen tube growth act as the primary
stimulus that initiates fruit development (Roth, 1977; Lee, 1987), but a second stimulus
is required to maintain fruit growth. This secondary stimulus emanates from the
developing seed and especially the endosperm (Luckwill, 1959; Lee, 1987) that starts
growing rapidly and immediately after fertilization, producing high levels of auxins. Other
phytohormones like gibberellins and cytokinins may also be involved (Lee, 1987). How
seeds influence fruit development is not clearly understood, but literature suggests that
whether a fruit aborts or matures depends on its ability to chemically inhibit
neighbouring fruits, or on its sink strength (Lee, 1987). High metabolic activity of the
developing embryo and endosperm and the consequent production of phytohormones
in these tissues create strong sinks for photosynthate and mineral nutrients. On the
other hand, fruits without seeds or with abortive seeds, constitute weak sinks that
ultimately fail to develop, unless the plant has the ability to synthesize phytohormones
like gibberellins (Coombe, 1960; Hayashi et al., 1968; Ivahori et al., 1968; Khalifa, 1966;
Lodhi et al., 1969) or auxins in the ovary wall itself (Roth, 1977). The avocado (Persea
Americana Mill.) may represent one such plant, as small amounts of various
phytohormone-like substances have been found in the mesocarp of the 'Fuerte'
avocado and 'seedless' fruits frequently occur on 'Fuerte' and 'Ettinger' trees.
These underdeveloped fruits, commonly referred to as 'cukes' because of their
cucumber-shaped appearance, are usually not entirely seedless, but contain an embryo
and endosperm during the earliest stages of fruit development. Reports on cuke
structure (Blumenfeld & Gazit, 1974; Tomer et al., 1980) describe the arrest of ovule
growth, the degeneration and ultimate destruction of the embryo sac and chalazal part
of the nucellus and integuments and the proliferation of the micropylar part of the outer
integument, so that this part of the seed coat sometimes covers the degenerated
tissues. Tomer et al., (1980) suggested that the micropylar part of the outer integument
keeps growing, because the avocado seed coat contains a high level of phytohormones
that may exert a strong sink effect for photosynthates. This suggestion does not explain
why the chalazal part of the same structure then aborts, together with the developing
embryo and endosperm, resulting in the formation of a cuke instead of a normal fruit. No
attempt was made in the above-mentioned studies to relate seed structure to ovule
structure. During the course of the present investigation it became clear that basic
characters of the avocado ovule and seed, such as the pachychalaza, hypostase and
multicellar archesporial tissue have so far not been taken into account in literature
dealing with avocado fruit development. As previous investigations were executed on
F.A.A.-preserved, paraffin-imbedded material, sectioned at 15µ, we reasoned that better
fixation and sectioning procedures and a more detailed study of seed abortion may
provide the necessary information for determining the causal effects of cuke formation
in the avocado.
MATERIAL AND METHODS
Developing cukes, varying between 0.75 cm and 3 cm in length, were collected at
Westfalia in the Duiwelskloof district during the first week of October 1991 from five
trees, that had previously been classified as consistently low producers (hereafter
referred to as E-type trees). The trees flower profusely, have earlier first dates for
flowering as well as a longer flowering period than consistently high producers of normal
fruits (hereafter referred to as A-type trees) and shed most developing fruits at an early
stage. The few fruits that are retained, develop into cukes. When normal fruits do occur,
they are not dispersed among the cukes, but occur on separate branches.
Cukes were fixed in Carnoy for 24 hours and preserved in 70% ethanol. Flowers
collected earlier during the season from the same trees had been treated accordingly.
One hundred fruits as well as one hundred open flowers from the abovementioned five
trees were dissected to determine the presence of seeds/ovules. For detailed
embryological studies, additional flowers and early post-fertilization stages of the
flowers were collected two weeks later from the same trees, fixed and stored in a
phosphate-buffered solution (pH 7.4) of 5 percent formaldehyde that contained 0.5
percent caffeine to improve the fixation of phenolic-containing cells (Mueller &
Greenwood, 1977). Carnoy and formaldehyde-fixed flowers and fruits were classified
according to age/size in seven stages and processed for ultimate serial sectioning in
glycol methacrylate (GMA) at 3-5µ, using conventional procedures. At least 15 series
were obtained from each stage. Perfectly median (Fig. 1) as well as transmedian
longitudinal sections of all stages were made from GMA-embedded material. Sections
were stained with PAS/toluidine blue, but some received special staining to test for
lipids, proteins, starch, lignin and cutin.
The vasculature of ovules was determined by clearing the phenolic-containing tissues of
75 ovules from three E-type trees in a 50% v/v aqueous solution of household bleach
(Javel) for 30 to 60 minutes, rinsing briefly in water and staining in a saturated solution
of phloroglucinol in a 20% (v/v) aqueous solution of HCI.
RESULTS AND DISCUSSION
Macromorphology of the ovule
Of the flowers 96% had pistils of apparently normal size and structure. In 4% of the
flowers the solitary ovule was borne externally on the ovary. Each of the normal pistils
contained a single, pendulous, anatropous, bitegmic and crassinucellate ovule. In the
extreme apical part of the locule, the ovule that fits tightly into the locule, is attached to
the ventral, abaxial surface of the carpel by a short, overarching funicle (Figs. 1 & 2) so
that the raphe is dorsal. The outer integument is multi-layered, not discernable as a
separate entity on the raphal side and slightly shorterthanthefour-layered inner
integument. The latter integument forms the micropylar canal and becomes broader in
this region (Fig. 3). A meristematic zone in the chalaza, i.e. the tissues below the place
of attachment of both integuments
(Figs. 1,2,4,5 & 6), is responsible for the formation of a pachychalazal seed where the
seed coat consists of a pachychalazal, highly vascularised basal part covering most of
the seed and a small, integumentary part around the micropyle.
Cuke formation
Cukes develop from female sterile, cryptically male flowers on consistently low yielding
'Fuerte' trees. A hypostase that has, as yet, not been reported for the avocado, is
present in the chalazal tissue of the mature ovule and aborting seed (Figs. 5 & 6). This
layer seems to play a rolé in the degeneration of the peripheral nucellar tissue and the
non-development of a intercalary meristem (Fig. 4) responsible for the extension of the
pachychalaza. The ultimate cause of cuke formation, however, seemingly lies in the
disturbance of the polarity of the primordial nucellar tissue. Additional embryo sacs and
non-functional megaspores or megaspore mother cells (Fig. 5) that develop in the
nucellus, effect the collapse of the chalazal region of the dominant embryo sac.
Degeneration of these embryo sacs and megaspores causes the formation of nucellar
cavities (Fig. 7) that isolate the embryo sac from the chalazal flow of nutrients. If
fertilization has already taken place, a globular embryo and a limited amount of
endosperm tissue are formed (Fig. 6). Because the endosperm is starved of nutrients,
the formation of this tissue is curtailed at an early stage and embryo development
ceases. A meristematic zone that initiates from the inner layers of the outer integument,
directly opposite the place where the vascular supply to the chalaza terminates (Figs.
5,6 & 8), causes abnormal growth in the outer integument. It is suggested that, due to
the absence of meristematic activity in the chalazal region of the embryo sac and the
non-developing pachychalaza, resources are redistributed towards the stronger sink,
i.e. the outer integument. Consequently, this part of the seed coat proliferates, while the
embryo sac and pachychalaza degenerate. In spite of the abortion of the seed, the
pericarp of the cuke continues to develop, possibly because sufficient phytohormones
are produced by the cell activity in the outer integument to support minimal pericarp
development. A crude model of the structure of the ovary and ovule as well as the
possible processes involved in fruit development and cuke formation, is presented in
Fig. 9.
CONCLUSIONS
Cukes develop from flowers that are unable to produce viable seeds, i.e. are female
sterile, because endosperm development is curtailed at an early stage and the embryo
stops growing. Female sterility has previously been reported for the avocado (Tomer &
Gottreich, 1978; Tomer, Gottreich & Gazit, 1976; Sedgley & Griffen, 1980). Due to the
limited number of ovules sectioned for this study, aid, as well as to the University of
Pretoria we cannot claim that all flowers on E-type trees are female sterile, but our
results have shown that cuke formation is an indication of the presence of female sterile
flowers on avocado trees.
ACKNOWLEDGEMENTS
We are indebted to the FRD and the Westfalia Estate (Pty) Ltd for financial aid, as well
as to the University of Pretoria for providing the infrastructure to execute this
investigation.
REFERENCES
BLUMENFELD, A., GAZIT, S. 1974 Development of seeded and seedless avocado
fruits. Journal of American Society for Horticultural Science 99: 442 - 448
COOMBE, B.G. 1960 Relationship of growth and development to changes in sugars,
auxins and gibberellins in fruit of seeded and seedless varieties of Vitis. Plant
Physiology 35: 241 - 250.
HAYASHI, F. NAITO, R., BUKOVAC, M.J., SELL, H.M. 1968. Occurrence of gibberellin
A3 in the parthenocarpic apple fruit. Plant Physiology 43: 448 - 450.
IVAHORI, S., WEAVER, R.J., POOL, R.M. 1968. Gibberellin-like activity in berries of
seeded and seedless Tokay grapes. Plant Physiology 43: 333 - 337.
KHALIFA, R.A. 1966. Gibberellin-like substances from the developing banana fruit.
Plant Physiology 41: 771 - 773.
LEE, T.D. 1987. Patterns of fruit and seed production. In: DOUST, J.L., DOUST, L.L.
(eds) Plant reproductive ecology. Patterns and strategies. Oxford University Press,
Oxford.
LODHI, F., BRADLY, M.V., CRANE, J.C. 1969. Auxins and gibberellin-like substances
in the parthenocarpic and non-parthenocarpic syconia of Ficus carica cv. King.
Plant Physiology 44: 555 - 561.
LUCKWILL, L.C. 1959. Fruit growth in relation to external and internal chemical stimuli.
In: RUDNICK, D. (ed) Cell, organism and milieu. Ronald Press, New York.
MUELLER, W.C., GREENWOOD, A.D. 1977. The ultrastructure of phenolic-containing
cells fixed with caffeine. Journal of Experimental Botany 29: 757 - 764
ROTH, I. 1977. Fruits of angiosperms. Borntraeger, Berlin.
SEDGLEY, M. .GRIFFEN, A.R. 1980. Sexual reproduction of tree crops. Academic
Press, New York.
TOMER, E., GOTTREICH, M. 1978. Abnormalities in avocado (Persea americana Mill.)
ovule development. Botanical Gazette 139: 81 - 86.
TOMER, E., GOTTREICH, M., GAZIT, S. 1976. Defective ovules in avocado cultivars.
Journal of American Society for Horticultural Science 101: 620 - 623.
TOMER, E., GAZIT, S., EISENSTEIN, D. 1980. Seedless fruit in 'Fuerte' and 'Ettinger'
avocado. Journal of American Society for Horticultural Science 105: 341 - 346.